Integrating Error Propagation Theory Into the FMEDA Framework (Robert Bosch GmbH)

Why It Matters

By providing measurable uncertainty bounds, the approach boosts the credibility of functional‑safety analyses and reduces dependence on expert intuition, accelerating compliance and risk‑based decision‑making in the automotive semiconductor market.

Key Takeaways

• Error propagation adds statistical confidence to FMEDA metrics
• SPFM and LFM now include uncertainty intervals
• Error Importance Identifier highlights dominant uncertainty sources
• Enhances ASIC verification for ISO 26262 compliance
• Reduces reliance on subjective expert judgment

Pulse Analysis

Functional safety engineers have long relied on FMEDA to translate component failure data into system‑level safety metrics such as SPFM and LFM. While these numbers guide design decisions, their accuracy hinges on estimated failure‑mode distributions and diagnostic coverage, which are often derived from limited test data or expert opinion. This opacity creates hidden risk, especially in automotive ASICs where ISO 26262 mandates rigorous proof of safety. The new Bosch paper tackles this gap by applying error propagation theory, turning point estimates into statistically bounded intervals that explicitly convey confidence levels.

The proposed methodology propagates uncertainties from input parameters through the FMEDA calculations, yielding upper‑ and lower‑bound values for SPFM and LFM. A novel Error Importance Identifier (EII) ranks the input variables that contribute most to overall metric variance, allowing teams to focus validation resources where they matter most. By quantifying the maximum deviation and presenting confidence intervals, the approach offers a transparent, data‑driven view of analysis quality, moving safety cases from qualitative assertions to quantitative evidence.

For the automotive semiconductor ecosystem, this development could streamline ISO 26262 certification pathways and reduce costly redesign cycles caused by ambiguous safety margins. Manufacturers can now demonstrate not only that a metric meets a target value, but also how tightly that value is bounded, satisfying auditors and regulators alike. As the industry pushes toward higher integration densities and autonomous driving functions, adopting uncertainty‑aware FMEDA practices may become a competitive differentiator, fostering greater trust in ASIC safety across the supply chain.